The common cold is generally a relatively mild disease, however significant, complications resulting from colds, such as otitis media, sinusitis and asthma exacerbations are common. Human rhinoviruses (HRV) cause up to 50% of all adult colds and 25% of colds in children (Bella and Rossmann, J Struct Biol. 128:69-74, 1999, and Sperber and Hayden, Antimicrob Agents Chemother. 32:409-19, 1988). The cost to society runs into billions of dollar per year. These small, nonenveloped RNA viruses represent a subgroup of picornavirus (Rueckert, Virology, pp. 507-548, eds. Fields, et al., Raven Press, Ltd. New York, 1990) X-ray crystallography of rhinovirus identified a capsid 360 Å in diameter (1 Å=0.1 nm) with icosahedral symmetry, constructed from sixty copies each of the viral coat proteins VP1, VP2, and VP3 (Rossmann, Nature 317:145-153, 1985). A surface depression or “canyon” on HRV was suggested as the receptor binding site (Colonno, et al., Proc Natl Acad Sci USA. 85:5449-5453, 1985; Rossmann, et al. Nature 317:145-153, 1985). Of the 102 characterized HRV serotypes, 91 (known as the major group) share as their receptor a cell surface glycoprotein known as intercellular adhesion molecule-1 (ICAM-1) (Greve, et al., Cell 56:839-847, 1989; Staunton, et al., Cell 56:849-853, 1989); the binding site is located within N-terminal domain 1 (Greve, et al., J Virol. 65:6015-6023, 1991; Staunton, et al., Cell 61:243-254, 1990).
ICAM-1 is a membrane protein with five extracellular domains, a hydrophobic transmembrane domain, and a short cytoplasmic domain. ICAM-1 is expressed on many cells important in immune and inflammatory responses, and is inducible on others (Casasnovas, et al. Proc Natl Acad Sci USA. 95:4134-9, 1998). ICAM-1 functions as a ligand for the leukocyte integrins LFA-1 and Mac-1 (Springer, Cell. 76:301-14, 1994; Staunton et al., Cell 61:243-254, 1990). On the cell surface, ICAM-1 is primarily a diner due to association of the transmembrane domains (Miller, et al., J Exp Med. 182:1231-41, 1995; Reilly, et al J Immunol. 155:529-32, 1995).
Recombinant, soluble forms of ICAM-1 (sICAM-1) consisting of the five extracellular domains were shown to be effective in blocking rhinovirus infection of human cells in vitro (Greve, et al., J Virol. 65:6015-6023, 1991; Marlin, et al., Nature. 344:70-2, 1990). Evaluation of sICAM-1 activity against a spectrum of laboratory strains and field isolates showed that all major strains of HRV are sensitive to sICAM-1. Minor strains, which do not use ICAM as a receptor, were unaffected by sICAM-1 (Crump et al., Antiviral Chem. Chemother. 4:323-327, 1993; Ohlin, et al., Antimicrob Agents Chemother. 38:1413-5, 1994).
The anti-viral activity of soluble ICAM-1 in vitro appears to be mediated by more than one mechanism. These mechanisms include competition with cell-surface ICAM-1 for binding sites, interference with virus entry or uncoating, and direct inactivation by premature release of viral RNA and formation of empty capsids (Arruda, et al., Antimicrob Agents Chemother. 36:1186-1191, 1992; Greve, et al., J Virol. 65:6015-6023, 1991; Marlin, et al., Nature 344:70-2, 1990; Martin et al., J Virol. 67:3561-8, 1993).
The host range of HRV is restricted to primates. A recent study showed that soluble ICAM-1 was effective in preventing rhinovirus infection in chimpanzees (Huguenel, et al., Am J Respir Crit Care Med. 155:1206-10, 1997). Although chimpanzees do not show clinical symptoms, infection was demonstrated by measuring seroconversion and virus shedding. A single dose of 10 mg of soluble ICAM-1 as an intranasal spray was effective at preventing infection by HRV-16 when co-administered with HRV, or when the virus was administered ten minutes later.
A human clinical trial with soluble ICAM-1 showed that it reduced the severity of experimental HRV colds (Turner, et al., JAMA 281:1797-804, 1999). In this trial a total of 196 subjects received either soluble ICAM-1 or placebo in various formulations. Some subjects were given soluble ICAM-1 or placebo starting seven hours before inoculation with HRV 39 and others were started twelve hours after virus inoculation. Medications were administered as either an intranasal solution or powder, given in six daily doses for seven days (a total of 4.4 mg per day). In this study, soluble ICAM-1 did not prevent infection, as measured by either virus isolation or seroconversion (infection rate of 92% for placebo-treated vs. 85% of soluble ICAM-1 treated). However, soluble ICAM-1 did have an impact on all measures of illness. The total symptom score was reduced by 45%, the proportion of subjects with clinical colds was reduced 23% and nasal mucus weight was reduced by 56%. There was not a significant difference between the use of powder or solution formulations, or between pre- and post-inoculation groups. Treatment with soluble ICAM-1 did not result in any adverse effects or evidence of absorption through the nasal mucosa. Also, there was no inhibition of the development of anti-HRV type-specific antibodies.
As discussed, ICAM-1 is dimeric on the cell surface. Martin et al., in J Virol. 67:3561-8, (1993) first proposed that multivalent binding to HRV by a multimeric soluble ICAM might result in a higher effective affinity, termed avidity, and thus facilitate uncoating of the virus. They constructed multivalent, ICAM-1/immunoglobulin molecules, postulating that these would be more effective than monovalent soluble ICAM-1 in neutralizing HRV and thus would have increased therapeutic utility. These ICAM-1/immunoglobulin molecules included ICAM-1 amino-terminal domains 1 and 2 fused to the hinge and constant domains of the heavy chains of IgA1 (IC1-2D/IgA), IgM (IC1-2D/IgM) and IgG1 (IC1-2D/IgG). In addition, five extracellular domains were fused to IgA1 (IC1-5D/IgA). These ICAM-1/immunoglobulin molecules were compared with soluble forms of ICAM-1 having two (sIC1-2D) and five (sIC1-5D) domains in assays of HRV binding, infectivity and conformation. The ICAM-1/IgA immunoglobulin (IC1-5D/IgA) was 200 times, and the ICAM-1/IgM immunoglobulin (IC1-2D/IgM) and ICAM-1/IgG immunoglobulin molecules (IC1-2D/IgG) were 25 and 10 times, more effective than soluble ICAM-1. These molecules were highly effective in inhibiting rhinovirus binding to cells and disrupting the conformation of the virus capsid. The ICAM-1/IgA immunoglobulin molecules were effective in the nanomolar concentration range. Comparison of IC1-2D/IgA and IC1-2D/IgG showed that the class of Ig constant region used had a large impact on efficacy.
A subsequent study compared the inhibitory activities of soluble ICAM-1 and IC1-5D/IgA against nine major HRV serotypes and a variant of HRV-39 selected for moderate resistance to soluble ICAM-1 (Crump, et al., Antimicrob Agents Chemother. 38:1425-7, 1993). IC1-5D/IgA was more potent than monomeric soluble ICAM-1 by 50 to 143 times on a weight basis and by 60 to 170 times on a molar basis against the standard serotypes. The HRV-39 variant was 38-fold more resistant to soluble ICAM-1 than the wild-type, and it was only 5-fold more resistant to IC1-5D/IgA. This is consistent with the hypothesis that virus escape from inhibition by multivalent molecules would be expected to occur at lower frequency than virus escape from inhibition by monomeric soluble receptor (Martin, et al., J Virol. 67:3561-8, 1993). An assay designed to measure viral inactivation showed that HRV-39 and HRV-13 were not directly inactivated to a significant extent by soluble ICAM-1 (<0.5 log10 reduction in infectivity). However, incubation with IC1-5D/IgA resulted in a reduction of infectivity of these same viruses by about 1.0 log10 (Crump, et al., Antimicrob Agents Chemother. 38:1425-7, 1994). Results by Martin et al. (J Virol. 67:3561-8, 1993) suggest that the greater the valence, the greater the effectiveness of the molecules. Dimeric and decameric forms of IC1-2/IgM were separable by sucrose gradient sedimentation. The decameric form was five times more effective than the dimeric form at blocking binding of HRV to HeLa cells.
The ICAM-1/immunoglobulin molecules that have been described suffer from several drawbacks, including the laborious production techniques and high costs associated with those production methods. In addition, the previously described ICAM-1/immunoglobulin molecules have limited stability as multimers in the harsh environment in which the molecule must inactivate rhinoviruses.
The immunoadhesins of the present invention have significant advantages over what has been described in the art. The immunoadhesins of the present invention that are, expressed in plants would be tetrameric, rather than only dimeric. Immunoadhesins having multiple binding sites have a higher effective affinity for the virus, thereby increasing the effectiveness of the immunoadhesin. In addition, the association of secretory component and immunoglobin, J chain with the immunoadhesin of the present invention increases the stability of the immunoadhesin in the mucosal environment (Corthesy, Biochem Soc Trans. 25:471-475, 1997). Secretory IgA, which is associated with secretory component, is the antibody isotype normally found in mucosal secretions, including milk and colostrum. Unlike other antibody isotypes, SIgA can pass through the gut with very little proteolytic degradation. It also is very stable in crude plant preparations at room temperature. A function of the secretory component appears to be to protect the antibody from the harsh environment of the mucosa (Paul, Fundamental Immunology, Raven Press, NY, Third Edition, pp. 303-304, 1993). Furthermore, the immunoadhesin of the present invention are significantly less expensive to produce in plants than in animal cell culture, and production in plants would make it safer for human use, since plants are not known to harbor any animal viruses.